What are the conditions under which civilians will accept neighbors who cooperated or collaborated with an insurgent group back into their community after conflict? What is the effect of different accountability and transitional justice mechanisms on reconciliation and social cohesion in post-conflict settings? And how does variation in the identity of a rebel collaborator (e.g. gender, age, or co-ethnicity) affect the prospects for his or her reintegration into a post-conflict society? Finally, how does the form of collaboration (e.g. marriage to a fighter, military recruitment, or logistical support) interact with these other characteristics to determine willingness to accept rebel collaborators back into a community? Mosul, an Iraqi city that was controlled and governed by the Islamic State (IS) for more than three years, is an ideal site in which to seek answers to these questions, which are crucial to the stabilization and reconstruction of war-torn societies around the world. Given IS’s control over the territory of Mosul, capture of the local economy, and harsh treatment of dissidents, it is unsurprising that civilians overwhelmingly cooperated with the group—whether for ideological, materialistic, or self-preservational reasons. The spectrum of cooperative behaviors included (1) compliance with mandatory policies such as the obligation to pay taxes, (2) social integration with the group through marriage to its members, (3) employment in IS’s civilian workforce (for example, as teachers, engineers, doctors, or cooks), and finally, (4) recruitment as fighters. The Iraqi government is currently in the process of prosecuting thousands of alleged IS collaborators under a counter-terrorism law that specifies a minimum sentence of 15 years, even for those who were employed in civilian roles and never carried weapons or received military training. Yet, the causal effects of highly punitive responses to terrorism on social cohesion, national security, and recidivism remain poorly understood. Through an experiment embedded in a household survey of 1,400 residents of Mosul, we identify the varying effects of collaborator identity, nature of rebel collaboration, and type of punishment on preferences for reintegration.